Nearly 2.3 million grandmothers have a grandchild under 18 years of age living in their home (Bryson and Casper, 1998). Most of these grandmothers are caregivers for their grandchildren either as the primary, custodial caregiver or as a supplemental caregiver. In both cases, the expected duration of caregiving is often indefinite, although many primary caregivers can expect to raise their grandchildren until adulthood. The health of these grandmothers is important because of their central role in the family. In addition, grandmother-headed families are at greater risk because many such grandmothers report high stress, 22 percent are over 65 years, 23 percent are poor, and about half are minorities, increasing their likelihood of experiencing various health problems and associated social, role and functional limitations that affect family well-being. Indeed, the limited data on the health of grandmother caregivers suggest that primary caregiver grandmothers in particular have more problems with depressive symptoms and physical health than non- caregivers or caregivers of other types, although much less is known about the well-being of the 1.4 million grandmothers who provide supplemental care to children in multi-generational homes (Musil, in press; Szinovacz, 1999). In addition, the links between the grandmother and family in terms of stress, health and family well-being have not been previously studied in families of grandmother caregivers. Therefore, the purpose of this study is to describe the longitudinal effects of grandmother caregiver status, family stress, resources and coping on grandmother and family will-being. Research questions for this study are: 1. Are there differences between primary caregiver, supplemental caregiver and non-caregiver grandmothers in (a) family stress, (b) general health, (c) physical health, (d) mental health, and (e) family well being at each of three times one year apart, T1, T2, and T3? 2. Do these differences remain after controlling for partner status and race? 3. Do social support and learned resourcefulness moderate the relationship between appraised stress and individual and family well-being? 4. Does the model adequately represent the relationships among the major study variables with individual and family health at each of the three time points? And 5. How do grandmothers who experience changes in family composition perceive these changes for themselves and their families? Answers to these questions will be sought in a four and one half year longitudinal study of grandmothers using a random sample of 375 women from urban, rural and suburban areas in Ohio. Data will be collected by mailed questionnaire each year for three time points. A theoretical model based on the Resiliency Model of Family Stress, Adjustment and Adaptation (McCubbin and McCubbin, 1993) will be tested using structural equation modeling to describe relationships among the major study variables, thereby allowing nurses and other health care professionals to identify factors that may moderate the effects of stress on grandmother and family will-being. In addition, narrative interviews with grandmothers who experience changes in their family composition over the study period will illuminate how these changes affect the grandmother and family. The innovations of this study are that it includes a comparison group of non-care giver grandmothers in a longitudinal design using a probability sample, it considers the functioning of both the grandmother and family, and it examines the moderating effects of support and learned resourcefulness on the stress-health relationship.